ORLEN Skylight accelerator

We are looking for innovative microbial solutions supporting anaerobic digestion processes to improve the efficiency and stability of biogas production and increase methane (CH₄) yield from the feedstocks used.

The goal of the challenge is to identify biopreparations or technologies based on appropriately selected microorganisms and microbial consortia that enhance the breakdown of proteins, fats, and difficult-to-degrade organic compounds. We are particularly interested in solutions that can be selected or tailored to specific groups of feedstocks and demonstrate high effectiveness under industrial operating conditions.

We are looking for solutions that:

• increase the degradation and utilization of feedstocks in the anaerobic digestion process, contribute to higher biogas yield and/or methane (CH4) content,
• support long-term stability of the digestion process,
• are compatible with microorganisms involved in methanogenic digestion,
• maintain their effectiveness under varying process conditions and feedstock characteristics, can be adapted to different types of feedstocks used in biogas plants,
• are suitable for industrial-scale deployment while maintaining economic viability.

An important part of the challenge will be the opportunity to conduct a pilot project and validate the solution’s impact on process parameters and biogas production efficiency under real-world biogas plant operating conditions.

We invite startups and technology companies with solutions mature enough to be tested in an industrial environment.

We are looking for an AI-powered solution for the automated extraction, cleansing, standardization and integration of process data from technical documentation, including P&ID diagrams, and production systems. The solution should leverage OCR and document analysis capabilities to identify process equipment, instrumentation, measurement systems and technical tags.

The solution should support automated or semi-automated data mapping across multiple sources, in particular:

• mapping Instrument Tags to PI Tags in AVEVA PI,
• standardizing identifiers and tags referring to the same assets,
• validating identifiers, such as Compound IDs,
• detecting missing data, duplicates and inconsistencies,
• recommending corrections and providing confidence scores for proposed mappings.

A key component of the solution should be a Human-in-the-Loop mechanism enabling subject-matter experts to review, validate and approve proposed mappings and corrections.

The expected outcome is a consistent, standardized and reliable data model ready for use across solutions and environments such as AVEVA PI Asset Framework, analytics platforms, Data Governance frameworks, Digital Twins and AI-based applications.

We are looking for an IT solution leveraging artificial intelligence to support the development of a centralized process knowledge base combining information contained in technical and process documentation with real operational data from production units.

The goal of the challenge is to create a tool that supports process engineers and domain experts in quickly searching, analyzing and validating information related to technological processes, while also enabling advanced analysis of the quality and behavior of operational data.

The solution should enable the collection and intelligent processing of documentation available in various formats, including process instructions, technical documentation, procedures, reports, analyses and other materials describing the operation of industrial installations. The system should analyze not only the current content of documents, but also their revision history, successive versions, amendments and modifications.

In particular, the solution should support:

• automatic comparison of successive document versions,
• identification and summarization of introduced changes,
• identification of potential inconsistencies, missing information or content requiring revision,
• searching for information and relationships across multiple documents,
• generating answers and analyses with clear references to source documents,
• supporting users in preparing and updating documentation based on accumulated knowledge.

Ultimately, the solution should be integrated with systems used within ORLEN so that current documents, version information and revision history can be retrieved automatically. At the pilot stage, batch processing of a selected set of documents provided to the solution will be acceptable.

A second key source of knowledge should be process data originating from the PI system. During the initial stage, the solution may operate on prepared data extracts covering selected PI Tags and time periods. Ultimately, integration with the source system through available APIs or dedicated integration mechanisms is expected.

For process data analysis, the solution should support automated data assessment and preparation for further use, including:

• identification of missing data, frozen signals, outliers, spikes, drift and other anomalies,
• assessment of data quality and reliability,
• analysis of relationships and correlations between selected signals,
• identification of data segments requiring expert review,
• application of data cleansing, correction or completion mechanisms without modifying the source data,
• generation of statistics and insights regarding the quality of analyzed data.

A key value of the solution should be its ability to combine knowledge contained in documentation with insights derived from the actual behavior of technological processes. Users should be able to analyze operational data in the context of documentation related to a specific installation, process or technical issue, while the system should use both information sources when answering questions and conducting analyses.

For example, a user should be able to investigate the revision history of a specific process or instruction, identify documentation related to a particular technological issue, and then compare this information with actual process data from a selected period.

Interaction with the system should be supported through an intuitive interface, including the ability to ask questions in natural language. AI-generated answers should be verifiable through clear references to the source documents and data used in the analysis.

The solution should incorporate a Human-in-the-Loop approach, enabling experts to validate, correct and supplement information generated by the system. Change tracking, user access management and full auditability of system operations are also important requirements.

The architecture of the solution remains open. It may leverage technologies such as vector databases, knowledge graphs, RAG/GraphRAG, language models, statistical algorithms, machine learning or hybrid approaches. The expected solution should, however, be modular and capable of further development and integration with systems operating within the ORLEN environment.

The expected outcome of the pilot is to demonstrate the feasibility of creating a single environment supporting process engineers in working with both technical documentation and process data. In the longer term, such a platform may become the foundation for a corporate process knowledge base and for further analytics and AI use cases across ORLEN’s Downstream operations.

We are looking for an IT solution that enables users to independently build, develop and maintain digital models of technological processes and production facilities, integrating process knowledge, equipment structures, operational data and relationships between process elements.

The solution should enable the creation of a semantic model of the production environment compliant with ISA-95 and DEXPI standards, as well as its visualization through plant maps, process diagrams, asset hierarchies and knowledge graphs. An important element of the solution should be integration with production systems, including AVEVA PI Asset Framework.

The solution should serve as a common knowledge and analytics layer supporting use cases such as:

• digital process and material balances,
• monitoring and analysis of process deviations,
• soft sensors,
• relationship analysis and “what-if” scenario modelling,
• predictive and optimization models,
• data quality improvement and contextualization,
• development of agent-based applications and AI solutions.

The pilot should include the creation of a digital model for a selected process, integration with operational data, interactive visualization and demonstration of at least one advanced use case.

Ultimately, the solution should be scalable and provide a foundation for the digital representation of industrial processes and the further development of advanced analytics and AI applications across the production environment.

We are looking for a solution enabling the integration and centralized monitoring of personal gas detectors used by employees in industrial environments. The goal of the challenge is to create a single environment for real-time monitoring of devices from different manufacturers and to improve the ability to respond rapidly to events that may pose a risk to personnel. Currently, different types of personal gas detectors are used across industrial facilities. These devices operate independently and may use different communication technologies. We are looking for a solution that enables the broadest possible integration of the existing gas detection infrastructure without requiring its replacement with devices from a single manufacturer.

We are looking for solutions that can enable:

• integration of different types and brands of personal gas detectors,
• communication with devices using available interfaces and data transmission technologies,
• centralized real-time monitoring of detector status and operating parameters,
• collection and visualization of information on alarms and detected hazards,
• location tracking of devices and personnel, where supported by the technical capabilities of the detectors and available infrastructure,
• visualization of information within a single monitoring environment, e.g. through a dashboard or facility map,
• recording of events and historical data for HSE analysis and preventive measures,
• generation of alerts supporting rapid response by safety personnel.

We are looking for hardware, software or hybrid solutions capable of providing a vendor-agnostic integration and monitoring layer for both existing and future gas detection devices.

We are looking for a solution that enables the digitalization and automation of the submission, verification and approval of HSE questionnaires and other documentation confirming compliance with safety requirements by external suppliers and contractors.

The solution should reduce manual workload by automatically verifying document completeness, formal correctness of submitted information and the validity of required certificates. The system should enable standard cases to be processed without the involvement of specialist teams, such as HSE personnel, while automatically directing cases requiring additional assessment to the appropriate experts.

We are looking for solutions that:

• enable electronic submission and management of required documentation,
• automatically verify document completeness, formal compliance and validity periods,
• support approval workflows involving contractors, procurement/requesting personnel and, where required, HSE teams,
• identify cases requiring additional expert review,
• provide process status monitoring, change history and full auditability,
• support reporting, document management and monitoring of document and certificate expiry dates.

We are looking for a solution that enables the digitalization and improvement of processes related to explosion safety management, Ex equipment, hazardous areas, and associated documentation and inspections. The system should provide a unified environment giving employees and authorized contractors quick access to up-to-date and reliable information on Ex equipment, installations and hazardous areas, while supporting operations, maintenance and safety inspection processes.

We are looking for solutions that:

• enable unique identification of Ex equipment, installations, assets and hazardous areas, e.g. using QR codes or other digital identifiers,
• provide a centralized register of Ex equipment and associated ATEX documentation,
• enable on-site access to equipment documentation and history directly at the point of operation,
• support the planning, execution and documentation of inspections and maintenance activities,
• monitor equipment status, inspection and maintenance schedules, and documentation validity,
• enable the recording and management of non-conformities and corrective actions,
• provide a complete change history and full data auditability.

We are looking for an analytical solution leveraging Artificial Intelligence (AI), Machine Learning (ML) and advanced computational algorithms to dynamically determine the maximum permissible fill level of road and rail tankers transporting dangerous goods, in full compliance with ADR/RID regulations.

The goal of the challenge is to develop a decision support system for hydrocarbon loading operators that

• automatically accounts for product physico-chemical properties (e.g., temperature, density, expansion coefficient),
• considers tanker parameters (nominal capacity, filling ratio, design type),
• generates clear and regulation-compliant recommendations in real time,
• minimizes operational errors during the loading process.

The solution should support safe, accurate and consistent loading operations while ensuring adherence to applicable transport safety regulations.

We are looking for a cost-effective technological solution enabling continuous or high-frequency determination of the calorific value of gas within the distribution network, particularly in the context of increasing shares of biomethane and hydrogen.

The goal of the challenge is to complement conventional gas chromatography measurements with sensors, measurement devices or IT-based solutions that enable more granular gas quality monitoring across multiple points in the network and support the operator in the safe integration of renewable gases.

We are particularly interested in solutions that improve visibility of gas quality across the distribution network without requiring the large-scale deployment of conventional gas chromatographs.

The proposed solution should be suitable for testing within real-world gas distribution infrastructure and for operation under conditions compliant with ATEX requirements.

We are looking for a technological solution enabling controlled testing of natural gas-hydrogen (H₂) blend odorization using THT (tetrahydrothiophene) and assessment of how variations in the gas matrix affect the selectivity and response characteristics of process analyzers.

The goal of the challenge is to evaluate the effectiveness of existing odorization methods in the context of the planned or tested introduction of hydrogen into gas networks.

The results should provide a reliable basis for assessing the performance of THT-based odorization under varying hydrogen concentrations and, ultimately, support the selection and procurement of an appropriate odorization system.

We are looking for innovative hydrogen compression technologies that can provide an alternative to conventional mechanical compressors, particularly for applications in hydrogen refueling infrastructure.

We are interested in solutions that have reached at least the technology readiness level of a demonstrator at semi-industrial or pilot scale, including technologies such as electrochemical compression, membrane-based compression, or other novel approaches.

We are looking for solutions that can deliver:

• improved safety of hydrogen compression,
• reduced operating costs,
• fewer compression stages,
• elimination or significant reduction of contamination risks, such as those associated with oils or lubricants,
• performance parameters suitable for real-world hydrogen applications.

The proposed technology should demonstrate the potential for further scale-up and deployment in commercial hydrogen infrastructure.

We are looking for personal, portable multi-gas detectors designed for operation in explosive (EX / ATEX) zones that, in addition to standard gas detection functions, provide precise real-time location tracking.

The goal of the challenge is to enhance the safety of personnel working in high-risk industrial environments by:

• accurately determining their location,
• enabling rapid response in emergency situations,
• supporting rescue and evacuation operations.

The solution must be certified for use in EX environments and operate reliably under harsh industrial conditions, where traditional positioning technologies (e.g., GPS, 5G) may be insufficient or unavailable.